Currently, a 3rd generation (3G) mobile communication system is mainly based on standards of wideband code division multiple access (WCDMA) and code division multiple access (CDMA) communication systems.
The universal mobile telecommunication system (UMTS) is a 3G mobile communication system adopting the WCDMA air interface technology, and is thus usually referred to as a WCDMA communication system. The structure of the UMTS is shown in FIG. 1, which includes a UE, a radio access network (RAN), and a core network (CN). The RAN is a UMTS terrestrial radio access network (UTRAN), and is capable of handling all functions related to radio services. The CN handles all voice calls and data connections in the UMTS, and realizes switching and routing functions with external networks. The CN is logically divided into a circuit switched (CS) domain and a packet switched (PS) domain.
The structure of the UTRAN is shown in FIG. 2, and the UTRAN includes one or more radio network subsystems (RNSs). Each RNS is formed by a radio network controller (RNC) and one or more NodeBs. The RNC is connected to the CN through an Iu interface, and the NodeB is connected to the RNC through an Iub interface. In the UTRAN, the RNCs are interconnected through an Iur interface. The RNCs can be classified into source RNC (SRNC) and target RNC (TRNC) for the UE. The RNC is adapted to allocate and control radio resources of the NodeB connected or related thereto. The NodeB transmits data from the Iub interface to the UE through a Uu interface or transmits data from the Uu interface to the RNC through the Iub interface. The NodeB also participates in some radio resource management.
The network architectures shown in FIGS. 1 and 2 are old versions before 3GPP Release 6. Considering the network competitive power in the future, a brand-new 3GPP evolved network architecture is developing to meet application requirements of the mobile communication network in the coming decade or even longer time. The evolved network architecture includes system architecture evolution (SAE) and long term evolution (LTE) of the access network, and an evolved access network is called an E-UTRAN. The evolved network aims to provide a low delay, high data rate, high system capacity and coverage, and low cost network completely based on the Internet protocol (IP). Since the evolved network is a brand-new network architecture, all the nodes, functions, and flows of the network architectures shown in FIGS. 1 and 2 change substantially.
Currently, 3GPP includes SAE and LTE. LTE aims to provide a low-cost network capable of reducing the delay, raising the user data rate, and improving the system capacity and coverage. The LTE adopts only PS domain services, and the bearer network is an IP bearer. FIG. 3 is a schematic architectural view of the current evolved network. The LTE RAN is an RAN of the evolved network. In the network, the nodes include, but are not limited to, eNodeBs or control plane servers (CPSs), so logically be treated as LTE-RAN entities. The mobility management entity/user plane entity (MME/UPE) is also a logic entity. The MME is adapted to store the mobility management context of the UE, such as the user ID, mobility status and tracking area (TA) information, and certificate the user. The UPE is adapted to terminate the downstream data of the UE in an idle state, trigger a paging, and store the context of the UE, for example, the IP address and routing information of the UE. The user plane anchor may not change during the session of the UE. Whether the MME/UPE is separated or not makes no difference to the method provided by the embodiment of the present invention.
When the UMTS is advancing toward the evolved network, a stage that the UMTS and evolved network coexist appears. As shown in FIG. 4, interfaces S3 and S4 are added between the UMTS and the evolved network. The interface S3 realizes the exchange of information carried by UEs in an idle and/or active state between 3GPP access systems. The interface S4 provides control and mobility support for user planes between the general packet radio service (GPRS) CN and 3GPP anchors.
Currently, in Section 4.5a.1 of 3GPP protocol TS23.236-630, a method of transferring all the UEs of a source CN element in a 3G or 2nd generation (2G) network to other CN entities is provided. The method is implemented in three stages. In the first stage, UEs initiating a periodic routing area update (RAU) within a preset time limit are transferred according to a method shown in FIG. 5, and the involved network entities include a UE, an access network, a source CN element, and a target CN element. The method includes the following steps.
In Step 501, after entering an idle state, the UE activates a periodic RAU timer.
In Step 502, when the periodic RAU timer experiences a time-out, the UE sends an RAU request to the source CN element.
In Step 503, when the source CN element receives the RAU request sent by the UE, if the source CN element has already received an operation maintenance notification, all the UE loads have to be transferred to other CN entities. Thus, the source CN element sends an RAU acceptance message to the UE, and the message carries a temporary mobile subscriber identity (TMSI) re-allocated to the UE and a minute periodic RAU timer value (for example, 4 seconds) re-allocated to the UE. The network resource identifier (NRI) field of the TMSI is set as Null.
In Step 504, the UE receives the RAU acceptance message, and activates a new periodic RAU timer.
In Step 505, when the periodic RAU timer activated in Step 504 experiences a time-out, the UE sends an RAU request carrying the TMSI allocated to the UE in Step 503. After the access network receives the RAU request, as the NRI in the TMSI carried by the request is set as Null, the access network employs a non-access layer node selection function to choose a target CN element for the UE, and forwards the RAU request to the target CN element.
In Step 506, the target CN element receives the RAU request, and sends an RAU acceptance message to the UE.
However, the method in FIG. 5 has the following defects. The method is only applicable to the circumstance of transferring all the UEs serving in the source CN element. In the method, the source CN element passively waits for the UE to send a periodic RAU, instead of initiating the transferring of the whole or a part of the UEs in service. Moreover, the method is only for transferring UEs within the same pool area. In addition, the method is limited to 2G or 3G networks, and is inapplicable to other networks like evolved networks, networks where both evolved networks and UMTS networks coexist, or post-SAE networks.
Currently, in the evolved network, in Section 7.13 of 3GPP protocol TR23.882-130, a reattaching method is provided to transfer a UE from a source MME/UPE to a target MME/UPE. As shown in FIG. 6, the involved network entities include a UE, an evolved access network (eRAN), a target MME/UPE, a source MME/UPE, an inter AS anchor (IASA), and a home subscriber server (HSS). The method includes the following steps.
In Step 601, due to factors such as overload, routes not being optimized, or the reception of an operation and maintenance (O&M) request, the source MME/UPE sends a reattach request to the UE.
In Step 602, the UE finds a selectable evolved network access system through the interaction with the eRAN, and selects the access system and network.
In Step 603, the UE sends an attach request to the selected target MME/UPE, and the request carries the registration information of the UE such as packet temporary mobile subscriber identity (P-TMSI). If the UE does not store the registration information, the request carries a permanent user ID.
As the network is shared, the attach request carries the information of the selected target MME/UPE, and the eRAN selects the target MME/UPE. The attach request may also carry information about default IP access bearer, for example, the IP address and access point name (APN) selected by the UE.
In Step 604, on receiving the attach request, if the request carries the registration information of the UE, the target MME/UPE determines the source MME/UPE according to the P-TMSI, and sends a message for obtaining a UE-related user information to the source MME/UPE.
In Step 605, the source MME/UPE sends the UE-related user information, for example, permanent user information, to the target MME/UPE.
In Step 606, the target MME/UPE authenticates the UE through the HSS.
In Step 607, the target MME/UPE sends a registration message to the HSS, indicating that the target MME/UPE is serving the UE.
In Step 608, the HSS indicates the source MME/UPE to delete the stored UE-related user information or to identify the absence of the UE.
In Step 609, the HSS sends a registration acknowledge message to the target MME/UPE. Meanwhile, user subscription data authorized to the default IP access bearer is sent at the same time, and strategy charging control information of the default IP access bearer is also sent to the target MME/UPE at the same time.
In Step 610, the target MME/UPE selects the IASA.
In Step 611, the IASA configures an IP layer with the confirmed IP address of the UE, so that a new user plane is established, and a default strategy charging rule is adopted to charge the UE.
In Step 612, the target MME/UPE provides the eRAN with an allocation of quality of service (QoS) of the default IP access bearer, for example, an upper limit of the data transmission rate.
In Step 613, the target MME/UPE accepts the attach request of the UE, allocates the P-TMSI to the UE, and sends to the UE an attached acknowledge message carrying the confirmed IP address of the UE.
In Step 614, the UE sends the attached acknowledge message to the target MME/UPE.
The method illustrated in FIG. 6 is only applicable to an evolved network, and does not specify how to transfer a UE between CN entities in a UMTS network or in the case when a UMTS network and an evolved network coexist. FIG. 6 also reveals a defect that the access network may pick up the source CN element when selecting a CN element if no preventive measure is taken.